Lung cancer is the leading cause of cancer related death in the United States and a vast majority of the cancers are causally linked to smoking. Gene expression profiles in early stage non-small cell lung cancer (NSCLC) reflect unique aspects of individual tumors and provide precise prognostic information (Potti A, NEJM 2006). In preliminary studies, we also identified gene expression signatures that can predict the in vitro and in vivo response to various chemotherapeutic agents (Potti A, Nature Medicine 2006), specifically cisplatin therapy (Hsu D et al, JCO 2007) and also define an approach to identify the state of critical regulatory pathways within tumors. This proposal aims to build upon a strategy that makes use of gene expression signatures that can predict the likelihood of clinical response to cisplatin therapy in patients, combined with signatures that define the deregulation of various oncogenic pathways in NSCLC. The goal of this work is an integrated approach to the challenge of personalized treatment of patients with NSCLC - by identifying those patients who are likely to respond to cisplatin chemotherapy and also those who will ultimately fail chemotherapy, combining this with ERCC1 status, and finally integrating information from oncogenic pathway analysis can refine the use of a current repertoire of therapeutics;leading to a rationale approach to the goal of delivering the right treatment to any given patient. We believe that this approach will allow us to truly characterize a given tumor for the probability of response to chemotherapy and the status of relevant oncogenic pathways and then match therapeutics to the individual patient with advanced NSCLC, rather than a broad population of patients. During the study period, we propose to conduct a proof of concept phase II clinical trial validating the benefit of a genomic- guided chemotherapeutic strategy in NSCLC and further characterize the application of genomic predictors of chemotherapeutic response. Data from both of the above aims will further strengthen the rationale of Specific Aim 3 - to characterize patterns of oncogenic pathway activation in patients resistant to chemotherapy and thus refine targeted therapeutic options in NSCLC. Such a strategy will add to the validity of an approach that matches the right drug with the right patient, a critical step before the implementation of genomic-guided strategies in large scale clinical trials in NSCLC. PUBLIC HEALTH RELEVANCE: The development of gene expression profiles as predictive measures of outcome has the potential to vastly improve our understanding of cancer biology and outcomes. Equally important are the goals of this study which are to build upon a strategy that makes use of gene expression patterns to develop profiles that can predict the likelihood of sensitivity to chemotherapy combined with signatures that define the deregulation of oncogenic pathways in non-small cell lung cancer (NSCLC) - the development of comprehensive gene expression profiles that can serve to select the most appropriate therapy for the individual patient offers the potential to increase efficiency of therapy and to identify new therapeutic opportunities for patients likely to be resistant to current standard of care treatment.